40 
Proceedings of Royal Society of Edinburgh. [sess. 
lower part of the chest plays in effecting the emptying and (by 
resiliency) the consequent filling of the lungs. It has seemed 
desirable, therefore, to supplement them by further experiments, 
haying for their object the exact determination of the amount of 
air exchanged, not only per respiratory movement, hut also per 
unit of time, a factor which was left out of account in 
the earlier experiments, hut one, nevertheless, of considerable 
importance. 
The apparatus which was used in the experiments referred to 
in the report consisted of a counterpoised bell-jar, filled with air 
and inverted over water ; to or from this the air of respiration 
was conducted from the mouthpiece (or mask) by a curved tube 
which passed through the water and opened into the bell-jar. 
When, therefore, air was drawn by the movement of inspiration 
from the bell-jar this sank in the water, and when air was forced 
into it by the movement of expiration it rose. These movements 
of the bell-jar were recorded upon a slowly moving blackened 
cylinder, and the diameter and corresponding cubic contents of 
the bell-jar being known, the amount of air exchange was found by 
measuring the ordinates of the curves described on the cylinder. 
The readings, however, must be looked upon as only approximate, 
because, firstly, the bell-jar which was used was only approximately 
cylindrical ; and secondly, because the counterpoised bell-jar 
acquired, with the somewhat rapid movements imparted to it, a 
swing of its own which must have affected the record. 
In order to obtain more accurate measure of the amount of air 
exchanged in respiration, the apparatus which was employed in 
these earlier experiments has been discarded, and we have used a 
carefully constructed graduated gasometer (spirometer), counter- 
poised on the principle devised by the late Dr W. Marcet to avoid 
the error which arises from the fact that the more a gasometer is 
raised out of the water in which it is inverted, the greater is the 
pressure exerted upon its contents. The air which is pumped 
out of the chest is alone measured, but it is clear that an equal 
amount must afterwards pass in to take its place. The air is 
respired through either a mask or mouthpiece. In practice the 
latter is found to be the more convenient, as less liable to 
accidental leakage. When it is used, the nostrils must be occluded 
